Vous êtes sur la page 1sur 11

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1352-2752.


QMRIJ 14,4

The power of qualitative research in the era of social media

Alan Branthwaite
Ignition Marketing Research, Nantwich, UK, and


Simon Patterson
QRi Consulting, London, UK
Purpose The electronic social media such as Twitter, Facebook, MySpace, etc. have become a major form of communication, and the expression of attitudes and opinions, for the general public. Recently, they have also become a source of data for market researchers. This paper aims to provide a critical look at the advantages and limitations of such an approach to understanding brand perceptions and attitudes in the market place. Although the social media provide a wealth of data for automated content analyses, this review questions the validity and reliability of this research approach, and concludes that social media monitoring (SMM) is a poor substitute for in-depth qualitative research which has many advantages and benets. Design/methodology/approach The paper presents a detailed, systematic comparison of various research approaches. These include well-established methods and recent inventions which are in use to explore and understand consumer behaviour and attitudes. Particular attention is given to the analysis of spontaneous consumer attitudes as expressed through the social media and also in qualitative research interviews. Findings This analysis concludes that there are three critical features which differentiate qualitative research (as practised in IDIs and group discussions) from SMM. These are: the direct, interactive dialogue or conversation between consumers and researchers; the facility to listen and attend to the (sometimes unspoken) underlying narrative which connects consumers needs and aspirations, personal goals and driving forces to behaviour and brand choice; and the dynamic, interactive characteristics of the interview that achieve a meeting of minds to produce a shared understanding. Philosophically, it is this conversation that gives qualitative research its validity and authenticity which makes it superior to SMM. Originality/value This review questions the validity and reliability of the SMM, and concludes that it is a poor substitute for in-depth qualitative research which has many advantages and benets. Keywords Social media monitoring, Qualitative research, Research methods, Automated content analysis, Social media Paper type Research paper

Qualitative Market Research: An International Journal Vol. 14 No. 4, 2011 pp. 430-440 q Emerald Group Publishing Limited 1352-2752 DOI 10.1108/13522751111163245

Introduction This paper looks at some current issues regarding the use of qualitative methods for consumer research, in comparison with other techniques which are being strongly advocated, such as social media monitoring (SMM): . What is the current status and usefulness of qualitative research and its value for businesses? . How does it compare with other methods that have emerged and developed recently? . Fundamentally, what are the dening assets and tools of qualitative research for assisting business decisions, and increasing brand performance?

When reading professional magazines in the consumer research industry, it would be easy to get an impression that qualitative research is under attack from newer methods such as SMM. We reject this idea, based on a careful evaluation of different marketing research methods, and assert the uniqueness of the qualitative approach and its enduring values for business decision making and brand development. If we look at the underlying characteristics of established consumer research methods, we see there are two differentiating factors which inuence the choice of techniques: (1) the method of data collection which is either interrogative (as in focus groups, IDIs and also classic questionnaire surveys) or observational and un-intrusive (such as ethnography or semiotics); and (2) the method of analysis that is either qualitative or quantitative in approach. Enter SMM This is a new approach to consumer research. Essentially, SMM is an observational, passive and quantitative approach (Figure 1) and is quite different to other consumer research. If anything, it shares aspects of ethnography and quantitative surveys, so it is fundamentally different to the qualitative research approach. In practice, SMM taps into social networking sites such as Facebook, Twitter and Blogs by scraping (unseen and unheard) spontaneous opinions about brands. These comments relating to brands (in the sense that they are juxtaposed in the same phrase or sentence as a recognised brand name) are analysed in a highly automated way, using computer software to code the value judgements inherent in the words used. There are some obvious problems relating to this method of data collection and analysis, not least in knowing exactly which brand is being referred to when it is referenced simply as it or them after its rst introduction. Key Selling Points for SMM have been [. . .] . It captures the words and thoughts of a generation absorbed in publicising their opinions to friends and the world around them minute by minute. . The medium has buzz and immediacy for spontaneous self-expression. It is a new, exciting culture that people want to be seen to be part of. Gone is Andy Warhols 15 minutes of fame now it is continuous and ongoing.
Qualitative Classic qualitativefocus groups, IDIs Interrogative Semiotics ethnography Data collection Analysis Observational (un-intrusive)

Power of qualitative research 431

Classic quantitative surveys

Social media monitoring Quantitative

Figure 1. The anatomy of consumer research techniques

QMRIJ 14,4


Data are analysed minute by minute using software to summarise what is being talked about and what is said about your brand. The iconic cultural signicance of social media is demonstrated by the many movies which touch upon it now. This provides a novel, innovatory connection to consumers by eavesdropping on their everyday, on-going blogging and tweeting. However, this has raised data protection issues and questions about its compatibility with European laws.

But, is there a risk that some brand managers and consumer researchers are also getting caught up in this buzz without fully understanding the nature of the research tool? Limitations of SMM These mainly surround the analysis of the textual information collected, and the cultural norms surrounding the transactions which take place on social media sites. Text analytic software for analysing comments needs improvement In a recent review in Research Magazine (Golden, 2011), it was recognised that text analysis software is good at picking-up references to brands from the massive volumes of online communication generated each day. However, opinions were divided on the ability to code the underlying sentiments as positive or negative. A large-scale validating programme of the different types of text analytic software has not been conducted, and some suppliers are in denial about the validity and accuracy of these techniques. One study quoted in this article found that the analysis was only 30 per cent accurate. The software was thought to be better at analysing tweets which are limited to 140 characters, than postings on Facebook or Blogs which are longer. Some analytic software also seeks to improve validity by taking into account the proximity of positive or negative words to the brand. Some improvements have been made in identifying re-tweets which otherwise duplicate the same messages around cyberspace to create over-estimations in the analysis. There is also the potential for campaign groups, or even unscrupulous manufacturers, to articially bombard the social media with positive or negative opinions. It is reported (The Times, 2011) that Facebook paid a leading public relations agency, Burson-Marsteller, to orchestrate a campaign involving placing negative stories about Google in newspapers and online. The scandal came to light after a blogger revealed that the PR company offered to help write pieces criticising Googles rival social networking site that is under development. Although few studies have been published, the general view is that, in order to be reliable, the analysis often requires human intervention to remove duplications and also check which product or brand was actually referred to, as this can be separated by other interjections in a Blog. It is generally agreed that irony or sarcasm is very difcult for machines to understand. In one study for the CBS news organisation (BuyQualityEBooks.com, 2010) it was found that: . machine analysis lacked sensitivity and coded the majority of remarks as passive; and . human analysis found a substantial majority were positive (Table I).

Doubts about the validity of the cyber arena in which blogging takes place A second major area of concern relates to the culture of social media and the relationship between it and the attitudes and behaviour in real life: . The unseen inuence of the intended audience to whom remarks are directed, on the self-presentation of bloggers.
People are using the internet to deliberately boost their social image. We may be forced to change our identity to free ourselves from the self-images we have created when we were young (Dr Eric Schmidt, CEO of Google).

Power of qualitative research 433

Bloggers adopt different persona, attitudes and behaviour depending on the blogging context. The unanswered question in the quote below is: which Sarah Blow goes shopping for what and when?

I have a different persona on Facebook which is associated with my business, and on Twitter, which is a very different me, and very transient (Sarah Blow, Founder of Girly Geekdom Dinners).

Do social media opinions represent the real everyday consumer inside the blogger?

The very thing thats drawn everybody to the internet is also allowing us to be rude, mean, bitter and all in the cloak of darkness. [. . .] Re-invention, projection, inventing your idealised self these are a huge part of what Facebook means to our current cultural moment [. . .] Anybody who uses the internet knows that you can sit in your room [. . .] and not be the person you dont like in real life. (Aaron Sorkin, Scriptwriter of The Social Network lm).

Social media exaggerate and transform everyday attitudes and personalities, enabling us to live in two (or more), separated worlds. This casts doubt on the content of Blogs as a reliable expression of brand afnity that will translate into purchasing. Are bloggers attitudes and opinions representative of real life actions? In an article for The New Yorker, Gladwell (2010) explains how social media create weak ties and transient relationships, where the expressions of attitudes and feelings are unreliable and inconsistent. He points, for example, to the difference between the highly effective civil rights movement of the 1960s which was built on the strength of an intimate network of friendships, shared values and experiences, and the Save Darfur Facebook page which had 1,282,339 members, whose combined fund raising amounted to 15 cents each! However, in contrast, once groups and movements have arisen, the social media can be an important tool for communication and coordination of actions, as seen recently in Egypt, Tunisia and Libya, and nding loved ones after the Tsunami in Japan.

Positive Machine Human 11 63

Passive 84 30

Negative 5 7

Table I. Comparison of brand opinions in blogs coded by machine or humans

QMRIJ 14,4


These views are supported by psychological research (Urista et al., 2009; Ellison, 2007; La Rose and Eastin, 2004). The social media life and opinions of bloggers are unlikely to be representative of the real life attitudes of the consumer inside the blogger, especially for women (Li, 2005). The construction of social network personalities found in these studies indicates: . bloggers produce content for the admiration of others to increase their social capital; . they use social media sites for relaxation, fun and as a way of releasing emotions (escapism) as well as letting off steam and off-loading as in a confessional; . it is a way of gaining approval and a special identity; . by making yourself outstanding and interesting, you build up a distinctive identity (or notoriety); . a notable distinction is made by social media users between real-world friends and online friends who are not taken at face-value or as having the same credibility; . women treat social media postings as an isolated distinct identity; and . men treat their postings as a magnied, mass-media expression of their views and opinions to make themselves popular in the blogosphere. The cultural context of social media users Research reveals that SM users oscillate between the two worlds of social media and everyday life. This is not unusual, in so far as we all adopt different roles and persona, depending on the external context and situations we are operating in. What is exceptional about social media users is the extent to which they seek to be in a world that is self-constructed and articial, rather than situationally and socially determined. This is seen most vividly in Second-Life types of web site which also attract numerous users. The Everyday World is [. . .]: . Pragmatic, ordered, constrained. . Rule orientated (external, negotiated, conventions). . Collaborative, cooperative to achieve personal, social, and emotional outcomes. . Long term. Evolves gradually by consent and social inuence. . EGO driven. The Social Media World is [. . .]: . Self-indulging. Impractical. . Immediate, short-term goals. . Supercial, transient. . Readily satised, not by practical outcomes and effects, but by creating imaginary scenarios and fantasies. . These attributes have little relevance to the economics of the market place, as you pay little for your own fantasies. . Primarily ID driven.

The nature and status of SMM Based on this understanding of social media, we have evaluated the strengths and weaknesses of SMM in relation to other techniques and its usefulness as a research tool. Here, we are not considering the use of the internet as a medium for delivering questionnaires or conducting interviews, but as a method of gathering (or scraping, as it is popularly known) spontaneous information on attitudes and opinions. We considered 36 characteristics (including positive and negative aspects) found in consumer research techniques. Here are some examples of the similarities and differences found between SMM and quantitative or qualitative research approaches: (1) Examples of similarities that SMM shares with quant. or qual. research [. . .] . Quantitative and SMM similarities: large samples; essentially numeric data; and difculties evaluating the meaning of respondents spontaneous responses. . Qualitative and SMM similarities: gathers spontaneous views and opinions; depends on rigorous, insightful semantic analyses; and freedom for respondents to set the agenda and produce spontaneous ideas. (2) Compared to quant. and qual. research, SMM differs on characteristics such as [. . .] . SMM and quantitative differences: SMM has imprecise sampling, and no rating scales. quantitative has controlled, standardised questioning; and the inability to present stimuli for concept or advertising development research. . SMM and qualitative differences larger samples than qualitative research; no risk of hot-housing choices and decisions; no direct contact interaction with consumers; not focussed on research objectives; lacks non-verbal feedback and subtlety; and no opportunity to ground information in real life or the market place. It was clear that these three methods share some attributes but the patterns are surprising (further details can be found in Branthwaite and Patterson, 2010). Overall, out of the 36 characteristics that these techniques were compared on: . SMM shares few similarities with either the positive or negative characteristics of quantitative or qualitative research (Table II). . SMM has more points of difference with the features of both quantitative, and especially qualitative research (Table III).

Power of qualitative research 435

QMRIJ 14,4

From a methodological point of view, it is clear that SMM is not a replica or substitute for these established research methods. SMM is a very different research approach especially compared to qualitative methodology, being dissimilar on 12 out of the 36 attributes we examined.


The distinguishing values of qualitative research Reading research magazines and conference presentations, one gets the impression that SMM is seen as some kind of threat to live, interactive qualitative research, whether it is done face to face, by telephone or via the internet, although there are few similarities in approach, as we have shown. Three key features make qualitative research a unique and invaluable tool in marketing to consumers: (1) A conversation a direct dialogue with consumers that is physically face to face, over the telephone or across Skype. (2) Active listening for the underlying dialogue the mental stance; frame of mind; reluctant, half-suppressed comments and admissions. (3) An interactive merging of minds (or rapport) to achieve insights and possibilities that can be extrapolated to marketing issues. Those three features of qualitative research stand in opposition to SMM, which is: . based on an instantaneous, static expression giving a supercial appreciation of the consumer at a single moment; . unable to probe and explore the gaps in what is said (the space between words as described by Gordon and Corr (1991) to get to the meaning behind the utterance; . a remote and vague understanding of the respondent; and . powerlessness to extrapolate ndings to new scenarios through what if questions. The signicance of these three key qualitative features I. The importance of a conversation with consumers The Philosopher and Psychologist, Rom Harre (Harre, 1979; Branthwaite, 1983) drew attention to the way in which every social situation is regulated by a set of implicit rules and standards (values) that we come to understand intuitively The Dinner Party; The Ofce; The good night out; The metro; The interview.
SMM similarities with Positive features 3 4 Negative features 2 2

Table II.

Quant Qual

SMM differences from Table III. Quant Qual

Positive features 5 12

Negative features 4 4

We get uncomfortable when we nd ourselves in situations where we are uncertain of these unspoken rules and standards of expected behaviour (Figure 2). Everyday conversations (Verbal Accounts) also function within the rules of each particular situation (which is why it is difcult to interview effectively in a foreign culture): . in some situations, exaggeration of the truth is permissible to create a more dramatic and exciting account; . other situations demand more frankness and openness, which often involves an exchange of condences and secrets to reassure each other of delity before the conversation can go into deeper admissions; and . to be an everyday conversation, qualitative interviews should be based on the principles and rules of Getting to Know You. This is why it is good practice for interviewers to clarify the ground rules, and put people at ease, although most are now familiar with the protocol (or rules) for focus groups and consumer interviews. Harres striking innovation in our understanding of conversations was that in the way social situations work conversations are dependable and credible as insights into peoples thoughts, ideas, feelings and actions unless there are grounds in the situation or the manner of the speaker for believing otherwise. This is particularly true for future intentions and actions, whereas past actions (in hindsight) often require more justication or rationalisation, particularly when they go against the rules (or norms) of a situation. II. The power of listening is the second asset of qualitative research Active listening is achieved through key interviewer skills in: . building an atmosphere and relationship with the interviewee, of trust and acceptance (positive regard); . cultivating intrigued curiosity about other peoples lives, and what they tell us; . scanning continuously for alternative meanings or ambiguities in the conversations that may cover-up hidden motives, pleasures or dissatisfactions; . acute awareness of non-verbal expressions of values, feelings, and the over(or under-) tones in attitudes;

Power of qualitative research 437


Rules (implicit) Actor 'roles' Actions practical/self-expressive Verbal accounts commentary Audience 'judges'

Figure 2.

QMRIJ 14,4


tuning-in to different layers of the ongoing, internal conversations taking place within the interviewees mind the public/private, serious/playful, closing-down on the too private revelations, although with a little sensitive probing they may well reveal; spotting ambiguities over the course of an interview that need exploring and clarication for deeper meanings, ideas or solutions; and triangulation to check for reliability in what is said by asking the same question in different ways.

Emphasis was directed again to the importance of sensitive listening at the recent International Journal of Market Research (IJMR) Research Methods Forum entitled Start listening, stop asking. In a brief commentary on the IJMR Forum, Langmaid (2010) gives an extraordinary, personal insight into what listening entails:
Listening has character, just as speaking does, but we seem not to know it so well. We are always listening from somewhere; we have a tone of listening, just as we have a tone of voice.

He suggests that while listening to others there is both a foreground which is what the speaker is saying, and a background which is the ongoing, internal commentary that the listener has as they reect on what they are listening to. This background commentary has three modes: (1) Listening for safety to manage the conversation to avoid embarrassment (even aggression), steering clear of making a fuss or upsetting anyone. (2) Listening about what the speaker is saying and what the listener thinks about the speakers meaning; a rolling assessment and opinions of them, so what is heard is my interpretation of the speaker. (3) Listening for closure and conrmation am I getting the answers, what I need to know? What am I learning and what can I do about what the speaker says to full my agenda? Langmaid suggests the important thing in creating insights is to focus on listening out for possibilities. Qualitative research is much more effective at this task, and has a wide range of techniques to improve listening and understanding of what is meant in conversation. III. Empathy the third feature in the qualitative armoury The third characteristic of qualitative research which differentiates it from SMM, is the hardest to explain, but it is a major part of the skill-set of good practitioners. Empathy is the power of entering into anothers personality and intuitively sharing their experiences. This is based on experiencing their world from their position (literally achieving an inter-view or merging of minds) to be able to think and feel like them. This is achieved through the skill and sensitivity of interviewers: . Skill in overcoming the constraints and articiality of the interview situation. Being able to focus the interviewee on the social situation in which real consumer behaviour takes place. This is very important in any research project, but particularly in developing concepts and communications.

Sensitivity in listening to detect the verbal and non-verbal cues which differentiate between the rhetoric (what people say because it is a socially accepted view or explanation) and the personal, individuals perspective, meanings and value.

Power of qualitative research 439

These features of qualitative research enable us to get inside the mind of others. Without face-to-face (or voice-to-voice) interaction, there is the remoteness that characterises SMM. Conclusions (1) The social media are a potentially interesting and useful source of information on social discourse and attitudes. It provides a commentary on social life as it happens. (2) However, by its very nature it is a difcult medium to understand and interpret accurately in terms of meanings, attitudes and motivations. (3) SMM has a contribution to make to consumer research in collecting information from a particular medium that is growing in its use. (4) However, by its very nature, SMM is a very different technique to qualitative research. (5) The marketing objectives it can effectively address are widely different and more limited in not [. . .] . linking the meaning of utterances to underlying perceptions, motives and complex consumer relationships with brands; . investigating the nuances of product and brand concepts; and . anticipating future developments and changes. Qualitative research is a unique approach which is very adaptable [. . .]: . it works by understanding the context and intention of what consumers tell us; . it is holistic in its understanding of consumer perceptions, motivations and the underlying causes behind their actions in the market place; . its insights reveal possibilities and ways of improving brands; . it avoids short-term or momentary reactions among consumers, and looks for the over-view and enduring impact for brands; and . it brings understanding by appreciating the background of consumer reactions and the underlying drivers in purchasing.
References Branthwaite, A. (1983), Situations and social actions, Journal of the Market Research Society, Vol. 25, pp. 19-38. Branthwaite, A. and Patterson, S. (2010), The vitality of qualitative research in the era of blogs and tweeting, Proceedings of ESOMAR Qualitative Conference, Barcelona. BuyQualityEBooks.com (2010), available at: www.buyqualityebooks.com/ebooks/sentiment360 Ellison, N. (2007), The benets of Facebook friends, Journal of Computer-Mediated Communication, Vol. 12 No. 3.

QMRIJ 14,4


Gladwell, M. (2010), Small change: why the revolution will not be tweeted, The New Yorker, 4 October. Golden, P. (2011), Write here, write now, Research, May. Gordon, W. and Corr, D. (1991), The space between words, Journal of the Market Research Society, Vol. 32 No. 3, pp. 409-34. Harre, R. (1979), Social Being, Blackwell, Oxford. Langmaid, R. (2010), Co-creating the future, International Journal of Marketing Research, Vol. 52, pp. 131-5. La Rose, R. and Eastin, M. (2004), A social cognitive theory of internet uses and gratications, Journal of Broadcasting & Electronic Media, September. Li, D. (2005), Why do you blog?, MA thesis, Marquette University, Wisconsin, WI. (The) Times (2011), Facebook hit Google with slur campaign, The Times, 13 May. Urista, M.A., Dong, Q. and Day, K.D. (2009), Explaining why young adults use MySpace and Facebook, Human Communication, Vol. 12, pp. 215-29. About the authors Alan Branthwaite is an experienced practitioner in qualitative research. He has conducted studies for major international companies in around 45 countries worldwide, sometimes involving up to 20 countries in the same cross-cultural project. He was a founding Director of Cooper Research & Marketing (CRAM) and an academic at the University of Keele. Dr Alan Branthwaite has also been Director of Qualitative Research and Development at Millward Brown, where he was awarded an Atticus prize on four occasions by WPP for original published thinking in marketing services related to qualitative research and advertising. Alan is a Chartered Psychologist, a Fellow of the British Psychological Society, and a member of ESOMAR and the MRS. He is currently Managing Director of Ignition Marketing Research Consultancy and Consultant Director to QRi Consulting. Alan Branthwaite is the corresponding author and can be contacted at: alan@ignition-mrc.co.uk Simon Patterson is CEO of QRi Consulting and Consumer Psychologist specializing in International Marketing Research & Consultancy, who has personally conducted studies in over 40 countries and has worked on a broad spectrum of strategic marketing and branding issues for blue chip clients around the world. He is also a highly experienced moderator having personally conducted well over 1,000 workshops and focus groups in the UK and the USA. Simon was formerly Managing Director at CRAM International where he trained and worked for over 25 years, working closely with his mentor Chartered Psychologist the late Peter Cooper. Simon is a Visiting Lecturer at Cambridge University. He has given many papers at industry conferences (including ESOMAR, MRS, AQR and QRCA). He is a member of the IOD, AQR, and QRCA, a full member of the MRS, and a member of both the British Psychology Society and Society of Consumer Psychologists.

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints